PrecipitateâCoacervate Transformation in ... - ACS Publications

Subscriber access provided by ORTA DOGU TEKNIK UNIVERSITESI KUTUPHANESI

Article

Precipitate - Coacervate Transformation in Polyelectrolyte-Mixed Micelle Systems Fatih Comert, Duy Nguyen, Marguerite Rushanan, Peker Milas, Amy Y. Xu, and Paul L Dubin J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.6b12895 • Publication Date (Web): 11 Apr 2017 Downloaded from http://pubs.acs.org on April 18, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Precipitate - Coacervate Transformation in Polyelectrolyte-Mixed Micelle Systems Fatih Comert †, Duy Nguyen †, Marguerite Rushanan †, Peker Milas ‡, Amy Y. Xu †, and Paul L. Dubin †* †

Department of Chemistry, University of Massachusetts Amherst, 710 North Pleasant Street, Amherst, MA 01003, United States

‡

Department of Physics, University of Massachusetts Amherst, 666 North Pleasant Street, Amherst, MA 01003, United States *Email: [email protected]

ABSTRACT The polycation/anionic-nonionic mixed micelle, poly(diallyldimethylammonium chloride)sodium dodecyl sulfate/Triton X-100 (PDADMAC-SDS/TX100), is a model polyelectrolytecolloid system in that the micellar mole fraction of SDS (Y) controls the micelle surface charge density, thus modulating the polyelectrolyte-colloid interaction.

The exquisite temperature

dependence of this system provides an important additional variable, controlling both liquid-liquid (L-L) and liquid-solid (L-S) phase separation, both of which are driven by the entropy of small ion release. In order to elucidate these transitions, we applied high-precision turbidimetry (± 1 ppt), isothermal titration calorimetry, and epifluorescence microscopy which demonstrates preservation of micelle structure under all conditions. The L-S region at large Y including precipitation displays a remarkable linear, inverse Y-dependence of the L-S transition temperature Ts. In sharp contrast, 1 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 23

the critical temperature for L-L coacervation Tφ, shows nearly symmetrical effects of positive and negative deviations in Y from the point of soluble complex neutrality, which is controlled in solution by the micelle charge and the number of micelles bound per polymer chain n (Zcomplex = Zpolymer + nZmicelle). In solid-like states, n no longer signifies the number of micelles bound per polymer chain, since the proximity of micelles inverts the host-guest relationship with each micelle binding multiple PE chains. This intimate binding goes hand-in-hand with the entropy of release of micelle-localized charge-compensating ions whose concentration depends on Y. These ions need not be released in L-L coacervation, but during L-S transition their displacement by PE accounts for the inverse dependence of Ts on micelle charge, Y.

INTRODUCTION Liquid-liquid (L-L) or liquid-solid (L-S) phase transitions has been observed and studied for the mixtures of polyelectrolytes (PE) with oppositely charged micelles.1-6 In these systems the role of surfactant monomers is negligible when surfactant concentrations (typically on the order of 50 mM) are far above the critical micelle concentration (cmc) of the mixed micelle (approximately 0.2 mM for SDS/TX100).7 The predominance of PE-micelle interactions then allow these systems to serve as models for PE-colloid systems in general, bridging theories for the interaction of PE’s with isotropic oppositely charge surfaces,8,9 with real protein-PE systems in which the charge and geometry of the colloid is highly anisotropic.10,11 The widespread interest in the properties and phase behavior of these systems is due to their numerous applications in different areas ranging from personal care to pharmaceutics.12-14 Depending on the type of application, either L-L (i.e. coacervation) or L-S (i.e. precipitation) phase separation might be desirable. However, for most

2 ACS Paragon Plus Environment

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


applications involving coacervation of PE-colloid, irreversible L-S phase separation (precipitation) is problematic. Thus it is necessary to illuminate the nature of the L-S transition in such systems and its relationship to coacervation. Although factors influencing L-L phase separation (coacervation) of PE-colloid systems have been extensively studied,15-19 comparable understanding of the L-S phase separation is lacking. A striking virtue of polyelectrolyte-mixed micelle systems is the ability to monotonically (and often reversibly) vary micelle surface charge density σ by way of the ionic surfactant mole ratio Y = [Anionic]/[Anionic]+[Nonionic]. This makes it possible to demonstrate the onset of soluble complex formation at Yc,20 thus providing the first experimental validation of the theoretical prediction of a critical surface charge density σc,21 for phase-transition-like PE adsorption on surfaces of opposite charge.8,9 The energy of binding polymer segments to the colloid surface exceeds thermal energy at σc ξ ~ κa

(1)

where σ is the surface charge density, κ is the reciprocal Debye length, and ξ is the dimensionless PE charge density.21 Yc, corresponding to σc in (1), thus depends on ionic strength and PE charge, but not on its molecular weight or PE:surfactant stoichiometry. Ionic strength can also play a role unrelated to screening, when the adsorbed PE displaces condensed or locally accumulated counterions from the colloid surface,8,9,22-25 e.g. at large Y (vide infra). Other “critical” values of Y appear in the turbidimetric titrations of PE-nonionic micelles with ionic surfactant,20 although, unlike Yc, these transitions are subject to mass action laws, since colloid:PE ratios control the colloid-PE complex charge.26 The resultant phase boundaries show five regions separated by well-defined transitions at Yc, Yφ1, Yφ2, Ys. At Yc, association of near-



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 23

charge neutral soluble complexes leads to abrupt L-L phase separation (coacervation) at Yφ1. After maximum turbidity is reached, further increase in Y (away from complex charge neutrality) causes redissolution of the coacervate at Yφ2. Further increase in Y may lead to liquid-solid (L-S) transition, Ys. Basic understanding of the L-S transition has been lacking, even though elucidation of these transitions can be relevant to corresponding transitions at pHc, pHφ, pHp observed for protein-PE systems.27-29 Kumar et al extensively studied the temperature dependence of the PDADMAC-SDS/TX100 system, primarily as a function of Y, with limited work on effects of PE MW, ionic strength, and polymer-sufractant stoichiometry.30 Temperature-dependence of turbidity at different values of Y showed inflection points designated as Tφ, and the dependence of Tφ on Y constituted a phase boundary separating L-L (coacervate) regions from single-phase (solution) conditions. A local minimum in Tφ(Y) was observed at Y*~ 0.35, a condition shown in separate studies to correspond to the point of soluble complex charge neutrality.20 That this was the condition most favorable for coacervation was also confirmed by a maximum in turbidity at Y* upon isothermal variation in Y. Nearly symmetric increases in Tφ upon either increase or decrease in Y clearly arose from deviations from complex neutrality. However, this “binodal” appearance of Tφ(Y) completely disappeared at Y > 0.45, where Tφ varied inversely and almost linearly with Y, and visible observations suggested transitions to more dense phases, sometimes particulate. These transitions are the subject of the current work. Aside from the limited observations in Kumar et al, the region of precipitate formation has not been extensively studied in other polyelectrolyte-colloid systems, although PE-protein systems can also show transitions to either coacervate or precipitate depending on pH.29 The polyelectrolyte-micelle system offers important advantages, including temperature as a variable, 4 ACS Paragon Plus Environment

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


with charge- and geometric isotropy much less individualized than proteins. However, coacervate and amorphous precipitate may be observed together in the region of large Y,30 leaving unanswered the question of whether coacervate can transform into solid. This question is especially challenging when both phases tend to be observed together. While precipitate has not been observed to transform into coacervate, a related transformation i.e. complete dissolution of precipitate followed by the formation of coacervate, has been reported in polyelectrolyte-protein systems.29 Our interest here is to understand an apparently well-defined boundary for the liquid-solid transition in the PDADMAC-SDS/TX100 system, and the peculiar apparent linearity of Tφ with respect to Y. Highprecision turbidimetry, microscopy and measurements of water content were used to understand the effect of Y on the reversibility of precipitates heated to different temperatures. Calorimetry was used to indirectly reveal the magnitude and type of entropic driving forces as a function of Y. The integrity of micelles in reversibly temperature-induced precipitates was demonstrated via epifluorescence microscopy.

EXPERIMENTAL Turbidimetry. Appropriate amounts of PDADMAC (Mn = 458 kDa), and TX100 were dissolved separately in 0.40 M NaCl, and the two solutions were mixed after filtration (0.22 µm PES (Millipore)) to form a solution 1.5 g/L and 20 mM in PDADMAC and TX-100, respectively. 10 ml of the PDADMAC-TX100 mixture was titrated with 60 mM SDS in 0.40 M NaCl solution using a 2.0 mL microburet (Gilmont). After each 0.05 ml addition of SDS, corresponding to values of the micellar mole fraction (Eq. 2),

Y

[ SDS ] [ SDS ]  [TX 100]

(2) 5

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 23

the transmittance (T) was measured at λ = 420 nm with a Brinkmann PC800 probe colorimeter equipped with a 2 cm path length (b = 2 cm) optical probe. Values of %T, typically stabilized within 15s, were recorded after 1 min. For convenience, we used 100-%T, which is linear with the turbidity τ = - log T/b. Titrations were done at room temperature except where noted otherwise. Temperature ramp measurements were performed in order to determine Tφ and Ts, the critical temperatures for liquid-liquid, and liquid-solid Ts phase separations, respectively. The determination of these as points of deviation from lines of zero slope is shown for example, by inset A of Figure 2A. Epifluorescence

Microscopy.

Microscope

chambers,

with

flow

channel

dimensions

approximately 0.1 mm deep, 3 mm wide, and 20 mm long, were formed from upper and lower coverslips (22 mm x 22, and 22mm x 30mm, respectively) after treatment with UVO Cleaner (Model 42, Jelight Company Inc.) to minimize background fluorescence. Channel depth corresponded to thickness (0.1 mm) of double-sided tape. Immediately after filling with PDADMAC-SDS/TX100 solution (Y =0.55) containing 0.002 g/L Nile Red, the channel was sealed on both ends with silicone grease. Images were acquired using a Nikon Eclipse Ti microscope with a Zyla 5.5 sCMOS camera, and “extra-long working distance” condenser and objective lenses (0.3 NA and Plan Fluor 40x, 0.6 NA, respectively) both from Nikon, Melville, NY). The exposure time was 50 ms and images were acquired at 500 ms/frame. Heating and cooling rates of at 1 °C/min were obtained with a temperature controller stage (mK2000, Instec).

Isothermal Titration Calorimetry. In these “Type 2 titrations”, mixed SDS/TX100 micelles prepared in 0.4M NaCl solution with 77.6 g/L TX100 at Y = 0.30, 0.40 and 0.55, were titrated into


Page 7 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1.44 g/L PDADMAC, also in 0.40 M NaCl at 25 C in the injection cell of a VP-ITC microcalorimeter (MicroCal Inc., Northampton, MA). Both micelle and polymer solutions were degassed under vacuum for 7 min prior to titration. Each titration comprised 29 10 μL- injections, 300 s apart, under stirring with a magnetic flea at 307 rpm. The heats of dilution for both the micelle and polymer solutions were measured and subtracted from the raw ITC data appropriately. Dry weight measurements. The necessary volume of SDS (60 mM) was added to PDADMAC (1.5 g/L)- TX-100 (20 mM) mixture to bring Y to 0.55, 0.6, and 0.68. Amorphous precipitates were separated from supernatant by centrifugation (14000 rpm for 10 min). After removing supernatant, the precipitate was air dried to remove surface water. Precipitate was dried to constant sample weight at 100 °C. RESULTS AND DISCUSSION Transitions induced by change in micelle charge (Y) at fixed temperature Progression through the phase transitions with increasing micelle surface charge density is shown by the turbidimetric titration results in Figures 1 (A) and (B). The two well-known and well-characterized critical points Yc = 0.23 and Yφ1 = 0.31 correspond to the micelle surface charge densities leading to sequential formation of soluble complexes and then coacervate. A third transition at Yφ2 designates the abrupt dissolution of coacervate to form a stable one-phase system in which the charge of soluble complexes is net-negative.20 The intermediate state of Y* = 0.34 displays a local turbidity maximum, corresponding to complex charge neutrality based on measurements below Tφ.30 The onset of liquid-solid phase separation is seen at Ys = 0.55 at T= 29 °C. Figure 1B shows the appearance of the system at Yφ1, Y*, Yφ2, and Ys and beyond. The timedependent images at Ys = 0.55 (taken at 30 s intervals) suggest that this transition might be



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 23

kinetically controlled, unlike the rapidly reversible transitions of entering and leaving the coacervation region which is centered around Y* = 0.34. The absence of time dependence in the L-L separation region suggests different mechanisms for coacervation vs. precipitation. A)

B)

Figure 1. (A) Turbidity of PDADMAC (Mn = 458 kDa, 1.5 g/L)-SDS/TX100 as a function of Y (micellar mole fraction of anionic surfactant) in 0.4 M NaCl at 27 ± 1 °C. Transition points Yc, Yφ1, and Ys (0.23, 0.31, 0.55) correspond respectively to initial formation of soluble complex, coacervate, and precipitate. Inset: determination of transition points, images at 0.45, where the inverse dependence Tφ on Y is remarkably linear (regression coefficient 0.99) confirming a previously qualitative trend observed in ref 30. This boundary appears to encompass the formation of both very dense coacervates and amorphous precipitates. The principal significance of the phase boundaries of Figure 2A is the clear recognition of two distinct processes, one purely L-L and the other encompassing also L-S separations. These two processes overlap in isothermal turbidimetric titrations, particularly at higher values of T 30 as suggested by the intersection of the dashed lines. The result of this overlap can be unresolved coacervation and precipitation in turbidimetric titrations (MS in preparation). Those extrapolations are also suggested by the results of turbidimetric titrations in which region II was diminished, thus more completely revealing Region I. The ability to diminish region II independently of region I occurs of different dependences on PE MW or salt concentration,30 and points to a difference in the underlying mechanisms of the two phenomena, central to the point of the current work.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

A)

Page 10 of 23

B)

Figure 2. A) Phase boundary of PDADMAC (Cp = 1.5 g/L)-SDS/TX100 obtained by heating at fixed Y; the measurement at Y = 0.68 is obtained by increasing Y at fixed T. Triangles refer to heating: coacervation ()Tφ; formation of solid dense phase ()Ts; coacervation and precipitation; () Ts. Squares refer to cooling: precipitation and coacervation ()Tdis; solid dense phase () Tdis. Inset A shows the determination of Tφ, Ts, and Tdis, see text. The shape of the “binodal region” centered around Y = 0.35 is drawn by analogy to one obtained in previous work with Cp = 1.0 g/L, shown in inset B. B) The relationship of turbidimetric titration at fixed T = 25 °C (lower) and the phase boundary (upper), see text.

While the phase boundaries can be obtained as noted by varying Y at constant T, the advantages of temperature as the independent variable include: non-invasive in situ incremental T adjustments compatible with microscopy; thermodynamic insights; and access to a second path leading to the same final conditions, for example Y = 0.55/T = 25 °C upper and lower plots of Figure 2B. Both methods lead to identical data points in region I (0.25 < Y < 0.40) or region II (0.45 < Y < 0.65), i.e. the system is path-independent. However, while the boundaries per se are independent of paths used to construct them, the precise properties of the resultant solid-like (amorphous precipitate) materials can vary, for example their reversibility. Remarkably, as will be discussed below, the temperatures of formation of more or less solid-like materials can reside on


Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


the same boundary, i.e. with simultaneous observation of fluid- and solid-like dense phases to be discussed below. The region 0.25 < Y < 0.45 is centered at Y* = 0.34. Figure 2B shows that Y* also corresponds to the minimum in Tφ, i.e. where coacervate forms most readily, because PE-bound micelles neutralize the charge of polycations leading to neutral soluble complexes. This “mirroring” of Figures 1A and 2A shown in Figure 2B, demonstrates the relationship between temperature and Y as agents of coacervation. Thus, an increase in polymer concentration (Inset B in Figure 2A) reduces the number of micelles bound per polymer chain n, thus requiring a higher charge per micelle, e.g. an increase in Y. For example, Y* increases from 0.30 to 0.34. On the other hand, complex electroneutrality appears to be irrelevant to phase separation in the linear Region II: even though complexes are becoming more net-negative with increasing Y, Tφ continues to fall. This Ydependence indicates a very different mechanism of phase separation: instead of near-symmetry about Y*, we observe a linear inverse dependence of Ts on Y at Y ≥ 0.45. To explain the relationship between Ts and Y at Y>0.40 (Region II) we propose that L-S phase separation, like L-L separation (coacervation) in region I, is also driven by the expulsion of counterions, but with a much greater loss of water, i.e. to ~10% vs. 80% in the coacervate. The linear boundary in region II can also encompass an additional fluid (referred to in ref. 30 as “region Va”) of high viscosity, easily obscured by subsequent or even simultaneous L-S separation (see Figure S1A). Thus, while counterion expulsion is responsible for phase separation in several regions, the state of the second phase is exquisitely sensitive to the degree of ion expulsion, which in turn depends on the nature of counterions to be expelled, i.e. bound or condensed. In the Manning approximation for the release of condensed counterions, the binding of a polycation segment of charge ZP, leads to the release of Z counterions, with an entropy gain proportional to 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 23

[Na+]m /[Na+]bulk, where [Na+]m is the effective local concentration of condensed counterions. The free energy of the transition in the vicinity of the phase boundary must be close to zero. As shown in Figure 3, the enthalpy change for micelle-binding is not merely small31 but in fact strongly positive, and has to be overcome by the entropy of release of counterion, namely SNa+. In view of the small spacing between PDADMAC charges (0.6 nm) relative to the spacing between micellar SO3- (1.3 nm, ref 30), the release of Na+ facilitates ion-pairing. Along the linear boundary of Figure 2A, the transition to dense phases can be accomplished by temperature increases of 0.5 is large enough for localization of counterions on spheres of comparable charge density.25 The release of such “condensed” counterions provides a larger entropy due to the large change in Na+ concentration accompanying their release into the bulk solvent from a dense layer near the surface of the micelle.33 An increase in the population of micelle-associated counterions might occur at large Y and so make an additional contribution to Sφ. While this effect may not be easily identified from the phase boundary, it might influence the nature of the dense phase formed at Tφ. The nature of the dense phase can in fact be deduced from its water content and the kinetics of its redissolution, as seen in Figure 4.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Y = 0.4

A

Page 14 of 23

Y = 0.55

B

Figure 4. Temperature-dependent turbidity for PDADMAC (Cp = 1.5 g/L) - SDS/TX100 at A) Region I, Y = 0.4; and B) Region II, Y = 0.55. B: L-S phase transition, with precipitation only upon after heating from 19.0 °C to 35 °C. Upon cooling, partial dissolution is seen at 21.4°C, and full dissolution at 7.8 °C. Heating curves are fully reproducible regardless of heating rate, but cooling curves are rate-dependent (see Figure 5B).

In contrast to Region I, e.g. at Y < 0.45, where the L-L transition (coacervation) is fully and rapidly reversible (Figure 4A), the cooling curves for Region II e.g. at Y > 0.45, (Figure 4B) are strongly displaced from the time-independent heating curve, in a manner that depends on the rate of cooling. We propose that this displacement of the rate-dependent cooling curve from the rateindependent heating curve is related to the nature of amorphous precipitate formed at the final Ts and Y. Thus, if coacervate alone is obtained upon heating (Y

PrecipitateâCoacervate Transformation in ... - ACS Publications

Recommend Documents

PrecipitateâCoacervate Transformation in ... - ACS Publications